Filters








126,132 Hits in 5.3 sec

Learning Multi-agent Search Strategies [chapter]

Malcolm J. A. Strens
2005 Lecture Notes in Computer Science  
Experimental results show the method is effective in a two-dimensional multi-pursuer evader searching task.  ...  We identify a specialised class of reinforcement learning problem in which the agent(s) have the goal of gathering information (identifying the hidden state).  ...  Policy parameterisation Our goal is for the learning agent to acquire effective control policies (searching behaviours).  ... 
doi:10.1007/978-3-540-32274-0_16 fatcat:imsbp6dryrgklbbkfnsoy6i5qq

Transfer Learning versus Multi-agent Learning regarding Distributed Decision-Making in Highway Traffic [article]

Mark Schutera, Niklas Goby, Dirk Neumann, Markus Reischl
2018 arXiv   pre-print
conduct multi-agent learning directly.  ...  In the first step traffic agents are trained by means of a deep reinforcement learning approach, being deployed inside an elitist evolutionary algorithm for hyperparameter search.  ...  Multi-agent Learning Within the multi-agent learning strategy, the agents are trained simultaneously without being aware of each other.  ... 
arXiv:1810.08515v1 fatcat:5geimpxlafddrhifkfadrdom7e

Learning to Collaborate: Multi-Scenario Ranking via Multi-Agent Reinforcement Learning [article]

Jun Feng, Heng Li, Minlie Huang, Shichen Liu, Wenwu Ou, Zhirong Wang, Xiaoyan Zhu
2018 arXiv   pre-print
In this paper, we formulate multi-scenario ranking as a fully cooperative, partially observable, multi-agent sequential decision problem.  ...  We propose a novel model named Multi-Agent Recurrent Deterministic Policy Gradient (MA-RDPG) which has a communication component for passing messages, several private actors (agents) for making actions  ...  , general multi-agent reinforcement learning model named Multi-Agent Recurrent Deterministic Policy Gradient.  ... 
arXiv:1809.06260v1 fatcat:jlc36jx2pjervchkm6mut65ly4

Statistics Based Q-learning Algorithm for Multi-Agent System and Application in RoboCup

Ya Xie, Zhonghua Huang
2014 Journal of Software  
This paper proposes statistic learning based Q-learning algorithm for Multi-Agent System, the agent can learn other agents' action policies through observing and counting the joint action, a concise but  ...  useful hypothesis is adopted to denote the optimal policies of other agents, the full joint probability of policies distribution guarantees the optimal action choice to the learning agent.  ...  INTRODUCTION Reinforcement Learning of Multi-Agent leads the traditional reinforcement learning technology into MAS (Multi-Agent System).  ... 
doi:10.4304/jsw.9.3.634-640 fatcat:vatuwbdqhrb7lpsttj33whdi24

Beyond Greedy Search: Tracking by Multi-Agent Reinforcement Learning-based Beam Search [article]

Xiao Wang, Zhe Chen, Jin Tang, Bin Luo, Dacheng Tao
2022 arXiv   pre-print
In this paper, we propose a novel multi-agent reinforcement learning based beam search strategy (termed BeamTracking) to address this issue.  ...  We take the target feature, proposal feature, and its response score as state, and also consider actions predicted by nearby agent, to train multi-agents to select their actions.  ...  tracking with multi-agent reinforcement learning based beam search strategy, termed BeamTracking.  ... 
arXiv:2205.09676v1 fatcat:fxgik65lx5gfrdggq5uzegziqm

Study and Application of Reinforcement Learning in Cooperative Strategy of the Robot Soccer Based on BDI Model

Guo Qi, Wu Bo-ying
2009 International Journal of Advanced Robotic Systems  
The dynamic cooperation model of multi-Agent is formed by combining reinforcement learning with BDI model.  ...  In this way, Agent can be ensured to search for each state-action as frequently as possible when it carries on choosing movements, so as to shorten the time of searching for the movement space so that  ...  In the research of multi-Agent, people have proposed the thinking model of multi-Agent--BDI model.  ... 
doi:10.5772/6795 fatcat:i3fgy5tkknghtcpw62eiklwuvq

Reinforcement Learning for Online Information Seeking [article]

Xiangyu Zhao and Long Xia and Jiliang Tang and Dawei Yin
2019 arXiv   pre-print
In this paper, we give an overview of deep reinforcement learning for search, recommendation, and online advertising from methodologies to applications, review representative algorithms, and discuss some  ...  Search, recommendation, and online advertising are the three most important information-providing mechanisms on the web.  ...  Policy Learning Reinforcement Learning is a class of learning problems in which the goal of an agent (or multi-agent) to find the policy to optimize some measures of its long-term performance.  ... 
arXiv:1812.07127v4 fatcat:pyc75g5hufcs5b3f75gonbkp24

Learning to reinforcement learn for Neural Architecture Search [article]

J. Gomez Robles, J. Vanschoren
2019 arXiv   pre-print
Reinforcement learning (RL) is a goal-oriented learning solution that has proven to be successful for Neural Architecture Search (NAS) on the CIFAR and ImageNet datasets.  ...  We also provide guidelines on the applicability of our framework in a more complex NAS setting by studying the progress of the agent when challenged to design multi-branch architectures.  ...  A particular search strategy for NAS is reinforcement learning (RL), where a so-called agent learns how to design neural networks by sampling architectures and using their numeric performance on a specific  ... 
arXiv:1911.03769v2 fatcat:77ysgcurorhqteb4ptcigjdbfi

Learning competitive pricing strategies by multi-agent reinforcement learning

Erich Kutschinski, Thomas Uthmann, Daniel Polani
2003 Journal of Economic Dynamics and Control  
Agents that perform this task can improve themselves by learning from past observations, possibly using reinforcement learning techniques.  ...  Co-learning of several adaptive agents against each other may lead to unforeseen results and increasingly dynamic behavior of the market.  ...  Di erent types of asynchronous multi-agent reinforcement learning (RL) will be used to determine optimal seller strategies.  ... 
doi:10.1016/s0165-1889(02)00122-7 fatcat:tiqagxrfcbgvlhcie5b2ryfcye

CBIR in Distributed Databases using a Multi-Agent System

David Picard, Matthieu Cord, Arnaud Revel
2006 2006 International Conference on Image Processing  
Our system, inspired by "ant-agents", uses labels provided by the user for learning both the searched category of images and the path to the most relevant databases.  ...  Hence, we introduce in this paper a new architecture for image retrieval in distributed image databases, based on multi-agent systems.  ...  , thus sparing both CPU and bandwidth • if a machine is down, machines still up can keep on processing the search task We introduce a new Multi-Agent System architecture dedicated to image retrieval on  ... 
doi:10.1109/icip.2006.313069 dblp:conf/icip/PicardCR06 fatcat:tjuw6mbttnetfj24232rqsrs4y

Hybrid Formation Control for Multi-Robot Hunters Based on Multi-Agent Deep Deterministic Policy Gradient

Oussama Hamed, Mohamed Hamlich
2021 The MENDEL Soft Computing journal : International Conference on Soft Computing MENDEL  
This paper proposed a hybrid formation control for hunting a dynamic target which is based on wolves' hunting behavior in order to search and capture the prey quickly and avoid its escape and Multi Agent  ...  Hunting a moving target with random behavior is an application that requires robust cooperation between several robots in the multi-robot system.  ...  Multi-Agent DDPG Reinforcement learning (RL) is a branch of machine learning that focuses on making progressive decisions.  ... 
doi:10.13164/mendel.2021.2.023 doaj:fedd95128b6e487bb1866d97dd6b9637 fatcat:vmwoxrlqh5hb7oafr6wmxqopwe

Mobile Robot Path Planning Based on Improved Q Learning Algorithm

Jiansheng Peng
2015 International Journal of Multimedia and Ubiquitous Engineering  
For path planning of mobile robot, the traditional Q learning algorithm easy to fall into local optimum, slow convergence etc. issues, this paper proposes a new greedy strategy, multi-target searching  ...  Don't need to create the environment model, the mobile robot from a single-target searching transform into multitarget searching an unknown environment, firstly, by the dynamic greedy strategy exploring  ...  Main Text This paper put forward a dynamic strategy and multi-objective search for improved Qlearning algorithm, the convergence rate of traditional Q-learning algorithm is slow, easy to fall into the  ... 
doi:10.14257/ijmue.2015.10.7.30 fatcat:qk67l5wzo5d77nnl3f37bze4re

Searching Collaborative Agents for Multi-plane Localization in 3D Ultrasound [article]

Yuhao Huang, Xin Yang, Rui Li, Jikuan Qian, Xiaoqiong Huang, Wenlong Shi, Haoran Dou, Chaoyu Chen, Yuanji Zhang, Huanjia Luo, Alejandro Frangi, Yi Xiong (+1 others)
2020 arXiv   pre-print
Second, we propose a novel collaborative strategy to strengthen agents' communication. Our strategy uses recurrent neural network (RNN) to learn the spatial relationship among SPs effectively.  ...  In this study, we propose a novel Multi-Agent Reinforcement Learning (MARL) framework to localize multiple uterine SPs in 3D US simultaneously. Our contribution is two-fold.  ...  GDAS based Multi-agent Searching In deep RL, the neural network architecture of the agent is crucial for good learning performance [7] .  ... 
arXiv:2007.15273v1 fatcat:oura5g5atvcrfiyxejkmf5va24

Implementation of modified SARSA learning technique in EMCAP

D. Ganesha, Vijayakumar Maragal Venkatamuni
2017 International Journal of Engineering & Technology  
State-Action-Reward-State-Action (SARSA) is an technique for learning a Markov decision process (MDP) strategy, used in for reinforcement learning int the field of artificial intelligence (AI) and machine  ...  Experiment are conducted to evaluate the performace for each agent individually. For result comparison among different agent, the same statistics were collected.  ...  A Multi agent learning address the problem domains, agents [8] involved is numerous. The search room considered is extraordinarily huge.  ... 
doi:10.14419/ijet.v7i1.5.9161 fatcat:tvnldbsj7ffnbns6sulu24vxt4

Using multi-agent systems for learning optimal policies for complex problems

Andreas Lommatzsch, Sahin Albayrak
2007 Proceedings of the 45th annual southeast regional conference on - ACM-SE 45  
For speeding up the learning process a multi-agent architecture is applied, that supports the simultaneous analysis of alternative strategies.  ...  We prove the advantages of our approach by successfully learning a control strategy for a model helicopter.  ...  In general, there are two possible approaches for dening a learning sequence in multi-agent systems: 1. All agents compute their strategies simultaneously.  ... 
doi:10.1145/1233341.1233385 dblp:conf/ACMse/LommatzschA07 fatcat:ql4wbfmn6zamhlg2ii35gk7k74
« Previous Showing results 1 — 15 out of 126,132 results